A Novel Weighted Distributed Clustering Algorithm for Mobile Ad hoc Networks

Clustering has been found to be an effective means of resource management for MANETs regarding network performance, routingprotocol design, Quality of Service and network modeling though it has yet to be refined to satisfy all the issues that might be faced bychoosing this approach. With this fundamental problem in mind, the paper proposes a new weighted distributed clustering algorithm(CBMD) for MANETs that pays special attention to the selection of a suitable node for the role of clusterhead and its performance,which is crucial to the effectiveness of the network stability, longevity and efficiency. The proposed clustering algorithm takes intoconsideration four elements in context, connectivity (C), residual battery power (B), average mobility (M) and distance (D) of the nodesto choose locally optimal clusterheads. Following independent calculation of each node's weight the neighbours require notification inorder to select a clusterhead that meets the highest calculated weight, thus making selection in a distributed manner that has currentinformation on the most recent status of neighbour nodes.VB.NET was used for reliable, verifiable and valid analysis of the simulationruns conducted to evaluate the performance of the proposed algorithm in terms of the number of clusters formed, reaffiliation count andnumbers of clusterhead changes. Results show that the proposed algorithm performs better than existing algorithms and is also tuneableto different kinds of network conditions